This invention relates to a device for removing nitrogen oxides from the exhaust of an internal combustion engine which is operated predominantly with an excess of air, as well as to a method for dosing a reducing-agent-containing additive into an exhaust line of an internal combustion engine which is operated predominantly with an excess of air.
German document DE 101 26 456 A1 discloses a device and a method for removing nitrogen oxides from the exhaust gas of lean-operated internal combustion engines by means of an ammonia-containing reducing agent, in which a nitrogen oxide reduction catalytic converter is used which is split into at least two parts which are separate from one another. Provided at the outlet side of each catalytic converter part is a sensor provided which measures the ammonia slip of the respective catalytic converter part. Metering of the reducing agent takes place in a regulated fashion on the basis of the measured ammonia slip. This permits differentiated evaluation of the entire catalytic converter volume, and the nitrogen oxide conversion can be improved in comparison with an integral measurement of an equally-large catalytic converter volume. Disadvantageous, however, are the costs for the sensors. In addition, the ammonia slip is a variable which merely indirectly characterizes the catalytic converter state. In addition, slip-regulated reducing agent metering proves to be difficult if the ammonia slip is to be completely avoided.
It is therefore an object of the invention to specify a device and a method which, in a simple manner, permit a further improved nitrogen oxide reduction for an internal combustion engine which is operated predominantly with an excess of air.
This object is achieved by means of a device having the features claimed.
According to the invention, filling level determining means for determining a reducing agent filling level of reducing agent which is stored in the first catalytic converter part and/or in the second catalytic converter part are provided, and the metering quantity of the additive can be set as a function of the reducing agent filling level, determined by the filling level determining means, of the first catalytic converter part and/or of the second catalytic converter part. It is preferably possible to set a mass-related or volume-related dosing rate of the additive. The reducing agent filling level is a highly informative variable for the evaluation of the catalytic converter state, and definitively determines the nitrogen oxide conversion and the reducing agent slip. The determination of the reducing agent filling level in the first catalytic converter part and/or in the second catalytic converter part permits a differentiated evaluation of the catalytic converter state, which makes it possible to optimally set the catalytic converter parts individually with regard to the nitrogen oxide conversion and the reducing agent slip. An optimization of nitrogen oxide conversion and reducing agent slip is consequently permitted overall.
Primarily considered as an additive for the exhaust gas is a substance containing ammonia or a substance suitable for the separation of ammonia such as urea, ammonium carbamate or the like. The additive can be metered to the exhaust gas in pure form or as an aqueous solution and/or as an aerosol-like mist at a metering point which is arranged in the exhaust line upstream of the first catalytic converter part. The metering point is preferably formed as a dosing valve with one or more nozzle openings. The nitrogen oxide reduction catalytic converter is preferably embodied as a classic SCR catalytic converter, based on tungsten oxide or vanadium pentoxide, or as a zeolitic catalytic converter. It is known that these catalytic converter types can store considerable quantities of ammonia, which acts as a reaction partner for a reduction of nitrogen oxides which are supplied with the exhaust gas. In this context, a reducing agent filling level is to be understood as the quantity of reducing agent stored in the catalytic converter bed. A relative variable is preferably used as a reducing agent filling level, which relative variable specifies the stored reducing agent quantity in relation to the maximum reducing agent quantity which can be stored under the present conditions.
The filling level determining means can comprise a sensor which is arranged in the catalytic converter bed, is sensitive to the reducing agent and is capable of measuring the stored reducing agent quantity. It is however preferable to determine the reducing agent filling level by calculation, so that it is possible to dispense with a filling level sensor of this type. The filling level determining means are therefore preferably designed as an electronic unit which can access stored data, preferably in the form of characteristic curves, characteristic diagrams and programs. A model-based calculative determination of the reducing agent filling level in the first and/or in the second catalytic converter part takes place on the basis of the data and additional input signals relating to definitive state variables of the nitrogen oxide reduction catalytic converter and of the exhaust gas, and also operating data of the internal combustion engine such as temperature, oxygen content of the exhaust gas, exhaust gas mass flow rate, internal combustion engine load and speed. Here, it is advantageous if the filling level determining means can communicate with an electronic internal combustion engine controller which is usually provided in any case, so that it is possible to access all operation-relevant variables. The filling level determining means can therefore also physically be an integral constituent part of the electronic internal combustion engine controller. The filling level determining means can however likewise be formed as a constituent part of the dosing unit.
Within the context of the invention, a dosing unit is to be understood as a control unit which carries out the setting of the metering or a dosing of the additive on the basis of the present information relating to the reducing agent filling level of the first and/or of the second catalytic converter part. Here, in order to set the metering quantity, the dosing unit controls corresponding actuating elements such as pumps, valves and the like. The metering unit is preferably embodied as a regulator which can adjust the reducing agent filling level of the first and/or of the second catalytic converter part. In this context, the reducing agent filling levels as initial variables, which the filling level determining means can generate, constitute regulating variables which can be adjusted to predefinable values on the basis of the additive metering quantity as an actuating variable. Here, metering quantity is preferably to be understood as the quantity of additive metered to the exhaust gas per unit time, with a dosing rate being generated here by averaging over a more or less extended time interval, so that one and the same dosing rate can be obtained both by means of pulsed on-off switching as well as by means of a variable adjustment within a continuous value range or by setting a technically predefined metering quantity within a certain timespan.
It is preferable to provide a separate regulating circuit for each catalytic converter part, so that the reducing agent filling levels of the first and of the second catalytic converter parts can be adjusted separately. It is particularly preferable to predefine reciprocal cross-relationships for the reducing agent filling levels, so that during the adjustment of the reducing agent filling level of one catalytic converter part, that of the second catalytic converter part is taken into account. It is particularly advantageous in this connection that merely a single metering point upstream of the first catalytic converter part is required in order to realize the invention.
As regards the catalytic converter parts, these can be constituent parts of a single-piece catalytic converter which can be considered as being split. In one embodiment of the invention, the first and the second catalytic converter parts are arranged spatially separate from one another. Here, it can be advantageous to arrange the catalytic converter parts comparatively far apart in the exhaust line, so that there is a comparatively large natural temperature drop of typically 50° C. or more between them in the case of a heated first catalytic converter part. The strong temperature dependency of the reducing agent storage capacity can be utilized in this way. This effect can be utilized particularly effectively by means of different catalytic converter materials for the first and second catalytic converter parts. Here, it is particularly advantageous if, in a further embodiment of the invention, the first catalytic converter part has a larger volume than the second catalytic converter part. As a result of the temperature drop along the exhaust gas path, the second catalytic converter part usually has a lower temperature than the first catalytic converter part. Consequently, for approximately equally high reducing agent filling levels, the second catalytic converter part can be designed to be smaller than the first catalytic converter part.
In a further embodiment of the invention, the reducing agent filling level of a respective catalytic converter part can be determined by means of summing and integration with respect to time of at least a conversion rate of reducing agent which is converted in the respective catalytic converter part, a desorption rate of reducing agent which is desorbed by the respective catalytic converter part and a supply rate of reducing agent which is supplied to the respective catalytic converter part with the exhaust gas. This permits a continuous determination of the present reducing agent filling level by means of a continuously carried out balancing of the variables which influence the filling level. In order to further improve the accuracy of the filling level determination, it is additionally possible to incorporate further factors in the balancing. It is additionally possible to incorporate reducing agent losses as a result of side reactions such as for example a direct oxidation of reducing agent. The influence of side reactions can also be contained in the conversion rate of the reducing agent.
In a further embodiment of the invention, a characteristic diagram set is provided for the conversion rate and for the desorption rate, which characteristic diagram set can be accessed by the filling level determining means during the determination of the reducing agent filling level. Here, the characteristic diagrams, or characteristic curves, expediently contain previously determined data for the respective catalytic converter. If further influential factors, such as for example a loss rate as a result of direct oxidation, are incorporated in addition to the conversion rate and the desorption rate, then separate characteristic diagrams which the filling level determining means can access can also be provided for the further influential factors. Losses as a result of side reactions can also be contained and incorporated in the characteristic diagram for the conversion rate.
In a further embodiment of the invention, for the reducing agent filling level of the first catalytic converter part, a first range with a first lower threshold value and a first upper threshold value can be predefined, and for the reducing agent filling level of the second catalytic converter part, a second range with a second lower threshold value and a second upper threshold value can be predefined. The dosing unit interacts with the filling level determining means in such a way as to permit an adjustment of the reducing agent filling level of the first catalytic converter part and/or of the second catalytic converter part into the respective range. The ranges for the reducing agent filling level are expediently predefined in advance in such a way as to generate an optimum nitrogen oxide conversion with a simultaneously minimal or negligible reducing agent slip. The ranges can be determined by means of suitably preliminary tests. The ranges for the reducing agent filling level can be different or identical for the first and the second catalytic converter parts.
In a further embodiment of the invention, for the nitrogen oxide reduction catalytic converter, an aging factor can be determined, and the first and/or second range for the reducing agent filling level can be predefined as a function of the aging factor. In this way, the ranges into which the reducing agent filling level is adjusted are adapted to the properties of the catalytic converter which vary in the course of time. Catalytic converter aging which occurs in the course of time can therefore be compensated, so that a high level of efficiency of the exhaust gas purification is ensured over long operating periods. If the first catalytic converter part is exposed to a higher temperature loading, it is advantageously possible by adapting the ranges to compensate the more sharply decreasing effectiveness of the first catalytic converter part. The separate consideration of different catalytic converter parts has an advantageous effect in this way. It can be provided to incorporate only aging caused by temperature loading. In this case, a weighted residence time at the respective temperature is expediently incorporated for determining the aging factor.
In a further embodiment of the invention, a first characteristic diagram set which is adapted for an unaged nitrogen oxide reduction catalytic converter and a second characteristic diagram set which is adapted for a definedly aged nitrogen oxide reduction catalytic converter are provided, with the first characteristic diagram set being assigned a first aging factor and the second characteristic diagram set being assigned a second aging factor. It is possible for the filling level determining means, during the determination of the reducing agent filling level, to carry out an interpolation between the data of the first characteristic diagram set and the second characteristic diagram set as a function of the present aging factor. This permits an appropriate incorporation of the present aging state of the first and/or second catalytic converter parts/part in a simple manner while avoiding an increased memory space requirement or processing expenditure. The aging factor is preferably defined between zero and one or 100%, with 100% aging constituting the maximum permissible aging, and the characteristic diagram sets corresponding to the aging states.
In a further embodiment of the invention, an exhaust gas sensor which is sensitive to nitrogen oxide and/or the reducing agent is arranged in the exhaust line downstream of the first catalytic converter part and/or of the second catalytic converter part. The exhaust gas sensor preferably serves for an alignment of the calculations carried out by the filling level determining means or for an alignment of the calculation model. For this purpose, it is possible to carry out a comparison between a calculated and a sensor-measured reducing agent desorption or nitrogen oxide concentration. This preferably takes place at predefinable or predefined operating points of the internal combustion engine or in a defined state of the catalytic converter, for example at every start of the internal combustion engine or before the beginning of a dosing.
In the method according to the invention, a reducing agent filling level of the first catalytic converter part and of the second catalytic converter part is determined and, as a function of the reducing agent filling level of the first catalytic converter part and/or of the second catalytic converter part, the reducing agent filling level of the first catalytic converter part or the reducing agent filling level of the second catalytic converter part is selected as a regulating variable for the regulation of the metering of the additive. By means of the determination, which takes place preferably in parallel, of the reducing agent filling levels both of the first and the second catalytic converter parts, the regulation of the metering can take place such that both catalytic converter parts in each case have predominantly their optimum reducing agent filling level, which is expediently adapted to the nitrogen oxide mass flow rates flowing in. This can be obtained by having the regulation of the additive metering oriented alternatively to the reducing agent filling level of the first or of the second catalytic converter part. Depending on the requirements, a switch is made between the reducing agent filling level of the first or of the second catalytic converter part as the present regulating variable for the metering. Here, the variables of the reducing agent filling levels are the definitive switching criteria. These can be flexibly predefinable or fixedly predefined. It is possible in this way to obtain overall an optimum nitrogen oxide conversion with at the same time negligible reducing agent slip.
In one embodiment of the method, the reducing agent filling level of a respective catalytic converter part is determined by means of summing and integration with respect to time of at least a conversion rate of reducing agent which is converted in the respective catalytic converter part, a desorption rate of reducing agent which is desorbed by the respective catalytic converter part and a supply rate of reducing agent which is supplied to the respective catalytic converter part with the exhaust gas. The conversion rate and desorption rate are expediently extracted from characteristic diagrams which are available for the respective operating state of the internal combustion engine or of a catalytic converter part. By means of the summing and integration of the variables which determine the respective filling level, the filling level can be determined continuously in a reliable fashion.
In a further embodiment of the method, the conversion of reducing agent, which is stored in the respective catalytic converter part, with nitrogen oxide and oxygen contained in the exhaust gas, is incorporated in the determination of the conversion rate. Particularly reliable determination of the present reducing agent conversion rate is permitted as a result of the additionally incorporated direct oxidation of the stored reducing agent with the oxygen present in the exhaust gas.
In a further embodiment of the method, the desorption rate of the first catalytic converter part is incorporated in the determination of the supply rate for the second catalytic converter part. This permits particularly reliable determination of the reducing agent filling level of the second catalytic converter part, so that the effectiveness or conversion rate can likewise be determined very reliably.
In a further embodiment of the method, for the reducing agent filling level of the first catalytic converter part, a first range with a first lower threshold value and a first upper threshold value is predefined, and the reducing agent filling level of the first catalytic converter part is selected as a regulating variable for the regulation of the metering of the additive if the reducing agent filling level of the first catalytic converter part lies below the first upper threshold value. With this approach, a target band for the reducing agent filling level of the first catalytic converter part is predefined. As long as the target band is not exceeded, the nitrogen oxide conversion rate can be further increased in the first catalytic converter part by means of further storage of reducing agent, without an impermissibly high degree of reducing agent slip occurring. If, therefore, in this case the reducing agent filling level of the first catalytic converter part is selected as a regulating variable for the metering, then the reducing agent filling level of the first catalytic converter part can be adjusted to its preferred value or into its preferred value range.
In a further embodiment of the method, for the reducing agent filling level of the first catalytic converter part, an upper limit value which lies above the first upper threshold value is predefined, and the reducing agent filling level of the second catalytic converter part is selected as a regulating variable for the regulation of the metering of the additive if the reducing agent filling level of the first catalytic converter part lies between the first upper threshold value and the upper limit value. With this approach, a switch is made to the reducing agent filling level of the second catalytic converter part as a regulating variable if the preferred target band for the reducing agent filling level of the first catalytic converter part is exceeded. Under these conditions, on the one hand, a sufficient nitrogen oxide conversion in the first catalytic converter part is permitted, and on the other hand, an appropriate reducing agent slip occurs, which is utilized for filling the reducing agent filling level of the second catalytic converter part. If, therefore, in this case the reducing agent filling level of the second catalytic converter part is selected as a regulating variable for the metering, then the reducing agent filling level of the second catalytic converter part can also be adjusted to its preferred value or into its preferred value range. At the same time, the reducing agent filling level of the first catalytic converter part is measured again in order to be able to react correspondingly, for example in the event of a further sharp rise.
In a further embodiment of the method, for the nitrogen oxide reduction catalytic converter, an aging factor which characterizes a temperature loading of the nitrogen oxide reduction catalytic converter is determined. This advantageously permits an adaptation of the preferred reducing agent filling levels of the first and/or of the second catalytic converter part to its respective aging state. Here, the temperature loading and the corresponding aging factor are preferably determined separately for the first and the second catalytic converter part.
In a further embodiment of the method, the first range and/or the upper limit value for the reducing agent filling level of the first catalytic converter part and/or the second range for the reducing agent filling level of the second catalytic converter part are therefore predefined as a function of the aging factor. It is therefore possible to react to different levels of aging of the catalytic converter parts.
Advantageous embodiments of the invention are illustrated in the drawings and are described below. Here, the features stated above and the features yet to be explained below can be used not only in the feature combination specified in each case but also in other combinations or individually without departing from the scope of the present invention.